Landslide susceptibility evaluation integrating weight of evidence model and InSAR results, west of Hubei Province, China

Abstract

Landslides are common geological hazards that take a heavy toll on human life and infrastructure every year, making it crucial to identify potential landslides. Landslide susceptibility (LS) mapping evaluates landslide proneness with the help of static data (e.g., topography, geology, and hydrology) that hardly change over decades; whereas the ground deformation map (GDM) produced by Interferometric Synthetic Aperture Radar (InSAR) works as a powerful reinforcement for determining landslide threats. The integration of the two could prompt the evaluation of the LS, yet the existing joint approaches exhibit certain defects. Therefore, a novel LS evaluation approach integrating the weight of evidence (WOE) model and InSAR results was proposed in this study. First, the GDM of the study area was obtained through InSAR process; meanwhile, an initial LS map was produced using the WOE model; subsequently, the LS map was coupled with GDM to develop a refined zoning and identification of landslide hazards. The proposed method was applied in Enshi, China, and achieved a higher AUC of 0.869. In addition, the accuracy of the susceptibility level of geological hazards found during field survey of the study area was improved, indicating that the LS map generated using the proposed method is more sensitive to landslide hazards and produces more accurate results.